ROAD OBJECT DETECTION WITH DEEP LEARNING
- 0 Collaborators
Road safety is a major concern today. Deep Automated driving and vehicle safety systems need object detection. It is important that object detection be accurate overall and robust to weather and environmental conditions and run in real-time. As a consequence of this approach, they require image. ...learn more
Project status: Under Development
Intel Technologies
oneAPI
Overview / Usage
Road safety is a major concern today. To overcome that realistic environmental issues it is required to develop suitable tools or GUI etc. Deep Automated driving and vehicle safety systems need object detection. It is important that object detection be accurate overall and robust to weather and environmental conditions and run in real-time. As a consequence of this approach, they require image processing algorithms to inspect the contents of images. This article compares the accuracy of a few major image processing algorithms: Deep Neural Networks (DNN) Ref [2], Region-based Fully Convolutional Networks (R-FCN), Mask Region-based Convolutional Neural Networks (Mask R-CNN), Single Shot Multi-Box Detector (SSD), RetinaNet, and You Only Look Once (YOLO) Ref [1]. Here we have taken care of Helmet-No Helmet detection and license plate photo capturing.
Methodology / Approach
INTRODUCTION
The automobile industries have developed rapidly since the first demonstration in the 1980s and the vehicle navigation and intelligence system have improved. However, the increase in road vehicles raises traffic congestion, road safety, pollution, etc. Autonomous driving is a challenging task; a small error in the system can lead to fatal accidents. Visual data play an essential role in enabling advanced driver-assistance systems in autonomous vehicles. The low cost and wide availability of vision-based sensors offer great potential to detect road incidents. Additionally, emerging autonomous vehicles use various sensors and deep learning methods to detect and classify four classes (such as a vehicle, pedestrian, traffic sign, and traffic light) to improve safety by monitoring the current road environment.
MOTIVATION
The motivation for studying road object detection is to understand the existing Object Detection Models, analyze the respective deep learning-based algorithm's performance for object detection under practical constraints. Some of the challenges that road object detection faces include:
● Come up with solutions to road traffic rule violations handing when there was no human monitoring
● The trade-off between speed and accuracy, as autonomous vehicles require real-time detection with high precision13.
● To understand and Hand-on experience of Object Detection algorithms implementation in Python.
● To contribute to society with new inventions to overcome practical problems which can be overcome with Object detection
● Collision avoidance systems have significantly decreased the injury and death rate in accidents. With the development of both software algorithms and hardware components, it is getting more precise for cameras to detect multiple objects in a real-time scenario using DNN.
The Intel Industrial Training initiative Unnati Program helps the students in getting the flavor of the Industrial View of the work planning, interaction, and guidance of the Intel Team and friendly competition with other college students.
DATA SOURCES
Studied and analyzed the Indian Driving Dataset IDD Lite for understanding the object detection procedure IDD (iiit.ac.in) and Google Images.
We have taken a few images with Mobile Camera on Hyderabad Road which has various objects like cars, trucks, helmets and no-helmet detections, and number plate capturing.
Developed code to take the input as images and object detection labels and create frames too with identified license plates of bikes without helmets.
Technologies Used
**PYTHON LIBRARIES USED IN THE PROGRAMS **
CV2 or CV: OPEN CV Python library of Python bindings designed to solve computer vision problems. OpenCV is a huge open-source library for computer vision, machine learning, and image processing.
Numpy: NumPy is a library for the Python programming language, adding support for large, multi-dimensional arrays and matrices, along with a large collection of high-level mathematical functions to operate on these arrays.
OS: The OS module in Python provides functions for interacting with the operating system.
Tensor flow: Keras is a neural network Application Programming Interface (API) for Python developed by Google that is tightly integrated with Tensor Flow, which is used to build machine learning models. Keras' models offer a simple, user-friendly way to define a neural network, which will then be built for you by Tensor Flow.
Imutils:
A series of convenience functions to make basic image processing functions such as translation, rotation, resizing, skeletonization, displaying Matplotlib images, sorting contours, detecting edges, and much more easier with OpenCV and Python 3.